feat: nav2 sensor fix OTA demo + cross-demo script regressions

.github/workflows/ci.yml

@@ -163,3 +163,93 @@ jobs:
         if: always()
         working-directory: demos/multi_ecu_aggregation
         run: docker compose --profile ci down
+
+  build-and-test-ota:
+    needs: lint
+    runs-on: ubuntu-24.04
+    steps:
+      - name: Show triggering source
+        if: github.event_name == 'repository_dispatch'
+        run: |
+          SHA="${{ github.event.client_payload.sha }}"
+          RUN_URL="${{ github.event.client_payload.run_url }}"
+          echo "## Triggered by ros2_medkit" >> "$GITHUB_STEP_SUMMARY"
+          echo "- Commit: \`${SHA:-unknown}\`" >> "$GITHUB_STEP_SUMMARY"
+          if [ -n "$RUN_URL" ]; then
+            echo "- Run: [View triggering run]($RUN_URL)" >> "$GITHUB_STEP_SUMMARY"
+          else
+            echo "- Run: (URL not provided)" >> "$GITHUB_STEP_SUMMARY"
+          fi
+
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Build and start OTA demo
+        working-directory: demos/ota_nav2_sensor_fix
+        run: docker compose up -d --build
+
+      - name: Run smoke tests
+        run: ./tests/smoke_test_ota.sh
+
+      - name: Show gateway logs on failure
+        if: failure()
+        working-directory: demos/ota_nav2_sensor_fix
+        run: docker compose logs gateway --tail=200
+
+      - name: Show update server logs on failure
+        if: failure()
+        working-directory: demos/ota_nav2_sensor_fix
+        run: docker compose logs ota_update_server --tail=200
+
+      - name: Teardown
+        if: always()
+        working-directory: demos/ota_nav2_sensor_fix
+        run: docker compose down
+
+  # Separate job from build-and-test-ota: this one publishes /goal_pose,
+  # waits for the controller to actually try to drive, asserts the
+  # reactive SCAN_PHANTOM_RETURN fault appears, and only then runs the
+  # OTA swap. Catches regressions in the demo narrative (broken_lidar
+  # subscribing /cmd_vel, fault_manager debounce, fixed_lidar not
+  # reporting). Slower than the API-only smoke job because it has to
+  # wait for nav2 lifecycle to settle and for /cmd_vel to actually fire,
+  # so it's split out and can fail in isolation without blocking the
+  # quick OTA-endpoint check.
+  ota-demo-narrative:
+    needs: lint
+    runs-on: ubuntu-24.04
+    steps:
+      - name: Show triggering source
+        if: github.event_name == 'repository_dispatch'
+        run: |
+          SHA="${{ github.event.client_payload.sha }}"
+          RUN_URL="${{ github.event.client_payload.run_url }}"
+          echo "## Triggered by ros2_medkit" >> "$GITHUB_STEP_SUMMARY"
+          echo "- Commit: \`${SHA:-unknown}\`" >> "$GITHUB_STEP_SUMMARY"
+          if [ -n "$RUN_URL" ]; then
+            echo "- Run: [View triggering run]($RUN_URL)" >> "$GITHUB_STEP_SUMMARY"
+          fi
+
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Build and start OTA demo
+        working-directory: demos/ota_nav2_sensor_fix
+        # docker compose up --build runs the multi-stage build for
+        # ota_update_server which produces the catalog + tarballs
+        # internally - no separate "build artifacts on host" step
+        # needed (and the host wouldn't have ros2_medkit_msgs anyway).
+        run: docker compose up -d --build
+
+      - name: Run demo narrative smoke
+        run: ./tests/smoke_test_demo_narrative.sh
+
+      - name: Show gateway logs on failure
+        if: failure()
+        working-directory: demos/ota_nav2_sensor_fix
+        run: docker compose logs gateway --tail=300
+
+      - name: Teardown
+        if: always()
+        working-directory: demos/ota_nav2_sensor_fix
+        run: docker compose down

README.md

@@ -49,6 +49,7 @@ All demos support:
 | [TurtleBot3 Integration](demos/turtlebot3_integration/) | Full ros2_medkit integration with TurtleBot3 and Nav2 | SOVD-compliant API, manifest-based discovery, fault management | ✅ Ready |
 | [MoveIt Pick-and-Place](demos/moveit_pick_place/) | Panda 7-DOF arm with MoveIt 2 manipulation and ros2_medkit | Planning fault detection, controller monitoring, joint limits | ✅ Ready |
 | [Multi-ECU Aggregation](demos/multi_ecu_aggregation/) | Multi-ECU peer aggregation with 3 ECUs (perception, planning, actuation), mDNS discovery, cross-ECU functions | Peer aggregation, mDNS discovery, cross-ECU functions | ✅ Ready |
+| [OTA over SOVD - nav2 sensor fix](demos/ota_nav2_sensor_fix/) | Dev-grade OTA plugin showing the SOVD `/updates` lifecycle - update a broken lidar node, install a new safety classifier, uninstall a deprecated package | SOVD-spec update / install / uninstall, native binary swap, fork+exec process management, Foxglove panel + curl scripts | ✅ Ready |
 
 ### Quick Start
 
@@ -150,6 +151,32 @@ cd demos/multi_ecu_aggregation
 - Unified SOVD-compliant REST API spanning all ECUs
 - Web UI for browsing aggregated entity hierarchy
 
+#### OTA over SOVD Demo (Dev-grade Update / Install / Uninstall)
+
+End-to-end demo of the SOVD `/updates` resource: a broken lidar node is
+swapped with a fixed version over HTTP, an extra safety classifier is
+installed from scratch, and a deprecated package is uninstalled - all
+without SSH, all spec-compliant.
+
+```bash
+cd demos/ota_nav2_sensor_fix
+./run-demo.sh                # build artifacts + bring up gateway/plugin/update server
+./check-demo.sh              # show registered updates + per-id status + live process state
+./trigger-update.sh          # broken_lidar -> fixed_lidar (the headline)
+./trigger-install.sh         # install obstacle_classifier_v2
+./trigger-uninstall.sh       # remove broken_lidar_legacy
+./stop-demo.sh
+```
+
+**Features:**
+
+- Dev-grade `ota_update_plugin` C++ gateway plugin (UpdateProvider + GatewayPlugin)
+- SOVD ISO 17978-3 compliant `/updates` resource: kind derived from
+  `updated_components` / `added_components` / `removed_components` metadata
+- Native binary swap + `fork+exec` process management (no containers, no signing)
+- Foxglove Studio panel mirrors the same SOVD client patterns as the web UI
+- Pairs with the [`ros2_medkit_foxglove_extension`](https://github.com/selfpatch/ros2_medkit_foxglove_extension) Updates panel
+
 ## Getting Started
 
 ### Prerequisites
@@ -209,9 +236,11 @@ Each demo has automated smoke tests that verify the gateway starts and the REST
 ./tests/smoke_test.sh              # Sensor diagnostics (full API coverage + fault injection + beacons)
 ./tests/smoke_test_turtlebot3.sh   # TurtleBot3 (discovery, data, operations, scripts, triggers, logs)
 ./tests/smoke_test_moveit.sh       # MoveIt pick-and-place (discovery, data, operations, scripts, triggers, logs)
+./tests/smoke_test_multi_ecu.sh    # Multi-ECU aggregation (per-ECU discovery + aggregated view)
+./tests/smoke_test_ota.sh          # OTA over SOVD (catalog, /updates spec shape, prepare/execute, process swap)
 ```
 
-CI runs all 4 demos in parallel - each job builds the Docker image, starts the container, and runs the smoke tests against it. See [CI workflow](.github/workflows/ci.yml).
+CI runs all demos in parallel - each job builds the Docker image, starts the container, and runs the smoke tests against it. See [CI workflow](.github/workflows/ci.yml).
 
 ## Related Projects
 

demos/moveit_pick_place/container_scripts/moveit-planning/inject-collision/script.bash

@@ -5,10 +5,13 @@
 
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 echo "Injecting COLLISION fault..."
 echo "   Spawning surprise obstacle in robot workspace"

demos/moveit_pick_place/container_scripts/moveit-planning/inject-planning-failure/script.bash

@@ -5,10 +5,13 @@
 
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 echo "Injecting PLANNING FAILURE fault..."
 echo "   Adding collision wall between pick and place positions"

demos/moveit_pick_place/container_scripts/moveit-planning/restore-normal/script.bash

@@ -4,10 +4,13 @@
 
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 GATEWAY_URL="${GATEWAY_URL:-http://localhost:8080}"
 API_BASE="${GATEWAY_URL}/api/v1"

demos/moveit_pick_place/setup-triggers.sh

@@ -2,8 +2,8 @@
 # Create fault-monitoring trigger for moveit pick-and-place demo
 # Alerts on any fault change reported by the manipulation monitor
 export ENTITY_TYPE="apps"
-# Uses ROS node name (underscore) - must match reporting_sources in FaultEvent
-export ENTITY_ID="manipulation_monitor"
+# Gateway normalises entity IDs with hyphens (matches the registered apps id).
+export ENTITY_ID="manipulation-monitor"
 export INJECT_HINT="./inject-planning-failure.sh"
 # shellcheck disable=SC1091
 source "$(cd "$(dirname "$0")" && pwd)/../../lib/setup-trigger.sh"

demos/moveit_pick_place/watch-triggers.sh

@@ -2,6 +2,6 @@
 # Watch trigger events for moveit pick-and-place demo
 # Connects to SSE stream and prints fault events in real time
 export ENTITY_TYPE="apps"
-export ENTITY_ID="manipulation_monitor"
+export ENTITY_ID="manipulation-monitor"
 # shellcheck disable=SC1091
 source "$(cd "$(dirname "$0")" && pwd)/../../lib/watch-trigger.sh" "$@"

demos/multi_ecu_aggregation/container_scripts/actuation-ecu/inject-gripper-jam/script.bash

@@ -2,10 +2,13 @@
 # Inject gripper jam - gripper controller stuck
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 ros2 param set /actuation/gripper_controller inject_jam true
 

demos/multi_ecu_aggregation/container_scripts/actuation-ecu/restore-normal/script.bash

@@ -2,10 +2,13 @@
 # Reset all actuation node parameters to defaults
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 ERRORS=0
 

demos/multi_ecu_aggregation/container_scripts/perception-ecu/inject-sensor-failure/script.bash

@@ -2,10 +2,13 @@
 # Inject LiDAR sensor failure - high failure probability
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 ros2 param set /perception/lidar_driver failure_probability 0.8
 

demos/multi_ecu_aggregation/container_scripts/perception-ecu/restore-normal/script.bash

@@ -2,10 +2,13 @@
 # Reset all perception node parameters to defaults
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 ERRORS=0
 

demos/multi_ecu_aggregation/container_scripts/planning-ecu/inject-planning-delay/script.bash

@@ -2,10 +2,13 @@
 # Inject path planning delay - 5000ms processing time
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 ros2 param set /planning/path_planner planning_delay_ms 5000
 

demos/multi_ecu_aggregation/container_scripts/planning-ecu/restore-normal/script.bash

@@ -2,10 +2,13 @@
 # Reset all planning node parameters to defaults
 set -eu
 
+# ROS setup.bash dereferences AMENT_TRACE_SETUP_FILES; relax nounset around it.
+set +u
 # shellcheck source=/dev/null
 source /opt/ros/jazzy/setup.bash
 # shellcheck source=/dev/null
 source /root/demo_ws/install/setup.bash
+set -u
 
 ERRORS=0